IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology

1. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 11 | Nov-2013, Available @ http://www.ijret.org 540
THROUGHPUT ANALYSIS OF ENERGY AWARE ROUTING PROTOCOL
FOR REAL-TIME LOAD DISTRIBUTION IN WIRELESS SENSOR
NETWORK (WSN)
A.T.I Fayeez1
, V.R Gannapathy2
, S. S. S Ranjit3
, S.K. Subramaniam4
, Ida S.Md Isa5
1, 2, 3, 4, 5
Faculty of Electronics & Computer Engineering, Uni.Teknikal Malaysia Melaka (UTeM), Melaka, Malaysia
fayeez@utem.edu.my
Abstract
Wireless sensor network (WSNs) are self-organized systems that depend on highly distributed and scattered low cost tiny devices.
These devices have some limitations such as processing capability, memory size, communication distance coverage and energy
capabilities. In order to maximize the autonomy of individual nodes and indirectly the lifetime of the network, most of the research
work is done on power saving techniques. Hence, we propose energy-aware load distribution technique that can provide an excellent
data transfer of packets from source to destination via hop by hop basis. Therefore, by making use of the cross-layer interactions
between the physical layer and the network layer thus leads to an improvement in energy efficiency of the entire network when
compared with other protocols and it also improves the response time in case of network change.
Keywords:- wireless sensor network, energy-aware, load distribution, power saving, cross layer interactions.
-------------------------------------------------------------------***----------------------------------------------------------------------
1. INTRODUCTION
The advances in micro-electro-mechanical system technologies
have made possible of embedding the system technology with
wireless connectivity which has low power energy usage hence
enabling sensing by using micro wireless sensors. Apart from
that, wireless communication and information processing is
also made possible. These power-efficient and inexpensive
sensor nodes works together to form a network that can be used
for monitoring the target region. Hence, WSNs sense and
transmit different elements of message about the
monitored/sensed environment such as temperature and
humidity by utilizing the cooperation of sensor nodes to sink
node which in turn processes the information and conveys it to
the user. WSNs have attracted serious interest and attention
from researchers from various field of study due to wide range
of applications such as military surveillance, health monitoring,
environment monitoring and many more.
Figure 1 depicts a standard wireless sensor network which
consists of several sensor nodes where one act as the sink
meaning that particular node maybe connected to a computer
for processing purpose of data received/sensed. However, there
lie some problems that require immediate attention such as
reliable data delivery, congestion control, energy consumption,
and data security to name a few. However, this study focus
solely on energy conservation method only which will be
discussed in later chapters.
Fig 1: Typical Wireless Sensor Network
Mostly, applications utilizing sensor nodes are powered by
batteries and must be able to work for several months or years
without need to recharge or replace. Therefore, such
expectation requires careful scheduling of the energy usage
especially when sensors are densely deployed (up to 20
nodes/m3 [1]), which causes severe, unwanted problems such
as scalability, redundancy, and radio channel contention [1].
Therefore, transmission of redundant data may occur due to
complex density whereby multiple sensor nodes can generate
and transmit about the same event to the sink node, causing
higher than normal energy consumption and hence a significant
reduction in network lifetime. Energy consumption in a sensor
node can be divided into three modules consisting firstly data

2. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 11 | Nov-2013, Available @ http://www.ijret.org 541
sensing module then data processing module and finally the
data transmission/reception module of which the energy
consumed for communication process is the most critical.
Hence, large amount of energy used can be saved by reducing
the amount of communication which can be done by
eliminating or aggregating redundant sensed data and using the
energy-saving link thus indirectly prolonging the lifetime of the
WSNs.
In this paper, we proposed a real time load distribution system
that takes into account the remaining battery level of the sensor
nodes before deciding which of the sensor nodes to be used for
creating the route towards destination or sink node. This
method introduces an efficient manner of energy management
system for WSN thus making full use of the available
remaining energy of each sensor nodes.
The remainder of this paper is organized as follows with
Section 2 reviews related works followed by Section 3 that
describes the system model and the motivation of our work.
Next, Section 4 presents the detailed design of the energy aware
protocol. Finally, Section 5 concludes the paper.
2. RELATED WORK
The main task of the sensor network is to relay the sensed data
gathered/sensed by sensor nodes to the base station. The WSN
administrator may use direct transmission by controlling the
transmission power. In this scenario, each sensor node within
the network will send sensed data directly to the sink node.
However, the node will be dead in short period of time due to
excessive energy consumption used up for delivering data if the
base station is far away from the sensor node. Therefore, some
algorithms were developed to overcome this problem.
One of the protocols proposed is Low-Energy Adaptive
Clustering Hierarchy (LEACH) by Heinzelman et al [2]. The
authors in [2] proposed and suggested LEACH which is an
alternative clustering-based algorithm. It assumes there exist a
unique base station outside the sensor network and all the
sensor nodes can communicate with this base station directly.
Meanwhile, LEACH only chooses a fraction p of all sensor
nodes to serve as cluster heads for the purpose of saving energy
where p is a parameter that must be determined before
deployment. The sensor nodes will then join the proper clusters
according to the respective signal strength from cluster heads.
The operation is divided into several rounds which ensure the
cluster head rotate in each round in order to share the energy
load. In LEACH, first phase is the cluster formation phase
which is followed by the process of cluster heads aggregating
the data received from cluster members and the aggregated data
is sent to base station by single hop communication. Hence, this
method is believed can effectively reduce the data needed to be
transmitted to the base station indirectly saving some amount of
power. LEACH's clustering algorithm assumes that sensor
nodes are homogenous and equal which in reality it is hard to
guarantee that. Moreover, according to the scale of the sensor
network, the optimum percentage of cluster heads must be
determined in advance. Therefore, LEACH cannot adapt to
such changes as the addition, removal, and transfer of sensor
nodes considerably affects the efficiency of data gathering.
Finally, a cluster head must broadcast its own existence to the
whole sensor network in cluster formation phase thus causing
another inefficient use of energy.
Meanwhile, S. Lindsey et al. proposed an algorithm called
PEGASIS [3] which has some relation to LEACH, The authors
noticed that the energy consumed by transmission circuits is
much higher when compared to that of amplifying circuits.
Therefore, PEGASIS make use of the GREED algorithm to
form the sensor nodes into a chain-like shape with intention to
reduce energy consumption of sensor nodes in the system. The
author’s findings show that PEGASIS performs better than
LEACH in the case of sink node that is far from sensor nodes.
On the other hand, Sh. Lee et al proposed a new clustering
algorithm called Congestion Detection and Avoidance (CODA)
[4] in order to relieve the imbalance of energy depletion caused
by different distances from the sink. CODA divides the whole
network into a few groups based on distance of the node to the
base station with each group has its own number of clusters and
member nodes. The further the distance to the sink node then
we require more creation of clusters in the case of single hop.
The paper outcome shows that the CODA have better
performance in terms of the network lifetime and the energy
loss than those protocols that use the same probability to the
entire network.
The authors in [5] proposed a hybrid with energy-efficient and
distributed clustering algorithm (HEED) which selects the
cluster head periodically according to two main parameters
which are the node residual energy and proximity of the node to
its neighbors. Low message overhead incurs in HEED thus
achieving quite uniform cluster head distribution within the
network. Meanwhile, the authors in [6] analyze the paper in
relation to extending the network lifetime by which
determining the optimal cluster size of WSN. They find that the
maximum lifetime or minimum energy consumption can be
achieved for certain optimal sizes of the cells. Hence, based on
their result, they propose a location-aware hybrid transmission
scheme that can minimize the energy consumption thus further
prolong the network lifetime.
3. ENERGY AWARE METRICS
As mention earlier, sensor nodes are low-powered devices thus
they have very limited energy hence it is necessary to maintain
the high energy level to achieve maximum network lifetime.
Loss of packets due to bit error or congestion can be considered
as common in WSN [7]. Hence, we can calculate the packet
loss ratio by using the equation below [7]:

3. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 11 | Nov-2013, Available @ http://www.ijret.org 542
(3.1)
Meanwhile, the number of packets successfully reach
destination can be determined as success rate of transmission
while packet latency is defined as time taken for a packet to
reach its destination from the moment of the packet creation.
Ratio of packet loss will result in energy loss per node and
affect the entire network hence enhancing the importance of
evaluating energy efficiency. Therefore, we assume that
dropped packets are directly proportional to energy wastage, so
the energy loss for each node can be measured by [7]:
(3.2)
On the other hand, the loss of energy for the entire network is
found by using the equation below [7]:
(3.3)
4. NETWORK SETUP
Fig 2: Layers involved
Figure 2 shows the layers involved in the development of the
project. As we can see, the process starts with request from
application layer which then invoke the transport layer to
provide reliable data transfer mechanism. This in turn will
invoke the network layer in order for best route selection. This
is where the research work was focused on whereby the best
route selection was determined by comparing remaining energy
level. The node with higher remaining energy level will be
chosen to form the route to destination. However, the routing
protocol will need the remaining battery power from physical
layer in order to compute the route selection. Hence, the need
for inter layer interactions for the purpose of sharing
information.
Fig 3: 4-node Network Setup
Figure 3 shows how the entire network is setup up where the
network is configured as so having 4 nodes where node A is the
source node and source D is the destination while node B and
node C will act as intermediate nodes In this research work, we
assume that node D is not within the vicinity of node A hence
no direct transmission from node A to node D.
Fig 4: Request and reply packet format
When the network is set up, the source node will have to
compute the best route to the destination when there are packets
to be transmitted. However, before route is determined, the
nodes will have to have prior knowledge of existing nodes
within its vicinity. Hence, all the nodes in the entire network
will invoke a request message as shown in Figure 4, containing
its address and request for remaining battery level. The packet
type column is to differentiate between request packet and reply
packet. This request message will be broadcasted meaning that
all the nodes within the vicinity of sender node will receive the
request message. Then, these nodes will reply to the sender
node with a reply packet (Figure 4) containing its address and
its remaining battery level. Now, all the nodes in the network
will have information about its neighboring nodes within its
vicinity and the remaining battery levels. Then only can a route
be determined using the remaining battery level information.
As an example, in this setup, all the nodes will invoke the
request message and reply to correspondent node. Then, node A
will compare between node B and node C for higher remaining
battery level. If node B is chosen as the one with higher level,
then node A will transmit the data packet to node B. Once
receiving the data packet, node B will check the destination

4. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 11 | Nov-2013, Available @ http://www.ijret.org 543
address of the packet and determine whether to forward the
packet or not. This process of inquiring neighborhood node’s
battery level for route creation is periodically done for better
energy or battery management. If it is done every time a packet
to be transmitted, then it will defeat the purpose of this research
work which is for energy efficiency.
5. ANALYSIS
Once the protocol has been developed, it is then programmed
into the sensor node for analysis purposes. The entire network
is shown in Figure 3 which consist of 4 sensor nodes with one
acting as source node (node A), two intermediate nodes (node
B,C) for hopping purpose and one as the base station or sink
node (node D). The test that was conducted covers the range of
1 meter to 15 meters as marked as d. 255 packets of dummy
data of “HELLO” message were generated during the analysis
and packet received by sink node were captured in order to
measure the performance of proposed routing protocol in terms
of the packet loss and energy loss. Table 1 below shows the
packet received and packet loss for distance under test for the
proposed routing while Figure 5 shows the captured packet and
packet loss occurring.
Table 1: Analysis of proposed protocol
Distance
(meter)
Packet
Received
By
Proposed
Protocol
Packet
Loss of
Proposed
Protocol
Packet
Received
by
Flooding
Algorithm
Packet
Loss of
Flooding
Algorithm
1 250 5 201 54
2 248 7 175 80
3 242 13 162 93
4 237 18 143 112
5 228 27 121 134
6 220 35 103 152
7 216 39 87 168
8 204 51 79 176
9 197 58 71 184
10 190 65 66 189
11 183 72 42 213
12 165 90 32 223
13 142 113 0 255
14 101 154 0 255
15 87 168 0 255
As we can see from both Table 1 and Figure 5 respectively,
there are packet drops occurring even the sensor node are as
near as 1 meter to each other. Though all the nodes collect their
neighbor information and calculate the best route as soon as the
entire network has been setup thus recognizing and reserving
the path for data forwarding, but this path might change due to
low battery of sensor node hence requiring the neighborhood
table to be updated and recalculated. Hence, these processing
may cause some packet loss to occur as can be seen in Figure 5.
Fig 5: Packet loss occurring
Fig 6: Packet Received
Fig 7: Packet Loss

5. IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
__________________________________________________________________________________________
Volume: 02 Issue: 11 | Nov-2013, Available @ http://www.ijret.org 544
Other than that, we also notice during the analysis that data
collision happens when nodes transmit information at the same
time to the node that requesting. Hence, the information is lost
thus preventing data to be forwarded to node with higher
remaining battery level. Meanwhile, we can notice increasing
packet loss when the distance of the sensor nodes were more
than 12 meters between each other. This might due to the
capability of the xbee module itself and surrounding factor too.
On the other hand, several issues were not taken into
consideration such as the multipath fading effects, reflection or
the optimum capability of xbee module transmission hence
packet loss bound to occur. We can conclude that proper
battery utilization or energy management is important in
wireless sensor network in order to prolong the network life
time.
6. FUTURE WORK
As stated before, some of the issues were not taken into
consideration such as simultaneous transmission, link quality
between nodes, delay and few more which hopefully can be
covered in future analysis. Apart from that, comparison with
similar protocols will also be conducted to compare the
performance of proposed energy-aware protocol.
ACKNOWLEDGEMENTS
The authors would like to take this opportunity to thanks those
who are contributes directly or indirectly in completion of this
article and also for their constructive comments. In addition, the
authors also would like to express our gratitude to Universiti
Teknikal Malaysia Melaka (UTeM) for the support and
encouragement.
REFERENCES
[1]. Estrin, D.; Girod, L.; Pottie, G. ivastava, M. Instrumenting
the World with Wireless Sensor Networks. Presented at the
International Conference on Acoustics, Speech, and Signal
Processing (ICASSP2001), Salt Lake City, Utah, May 2001.
[2]. Heinzelman, W.R.; Chandrakasan, A.; Balakrishnan, H.
Energy-efficient communication protocol for wireless
microsensor networks. Presented at the Proceedings of the
Hawaii International Conference on System Sciences, Maui,
Hawaii, Jan 2000.
[3]. Lindsey, S.; Raghavendra, C. PEGASIS: Power-Efficient
gathering in sensor information systems. Presented at the
Proceeding of IEEE Aerospace Conference, 2002; pp. 1125-
1130.
[4]. Lee, S.H.; Yoo, J.J.; Chung, T.C. Distance-based Energy
Efficient Clustering for Wireless Sensor Networks. Presented at
the Proceedings of the 29th Annual IEEE international
Conference on Local Computer Networks (LCN'04), 2004.
[5]. Younis, O.; Fahmy, S. HEED: A Hybrid, Energy-Efficient,
Distributed Clustering Approach for Ad Hoc Sensor Networks.
IEEE Trans. Mob. Comput. 2004, 3, 366-379.
[6]. Xue, Q.; Ganz, A. Maximizing Sensor Network Lifetime:
Analysis and Design Guides. Proceedings of MILCOMOctober
2004
[7]. M. A. Rahman, A. E. Saddik and W. Gueaieb, “Wireless
Sensor Network Transport Layer: State of the Art”, Sensors,
Springer-Verlang Berlin Heidelberg, 2008.
[8]. Ida S.Md Isa, Sharifah H.S. Ariffin, N.Fisal,
N.M.AbdulLatiff, N.Latif A.Shaari, A.T.I Fayeez, “Multi Hop
Transmission in IEEE 802.11 Low Rate Ad Hoc Network
Using ARP-Route”, Journal of Telecommunication, Electronic
and Computer Engineering (2180-1843), Vol.5 No.1, Jan-June
2013.
[9]. V.R Gannapathy, M.R Ahmad, M.K Suaidi, M.S Johal,
“Performance Analysis of Concurrent Transmission With
Reducing Handshakes in Multi-hop Wireless Mesh Networks
(WMNs)”, JTEC 2009 Vol 1 No.1 (Jul-Dec 2009).